Study setting

This qualitative study exploring mental health service integration was embedded and conducted within Strengthening the Blood Pressure Care and Treatment cascade for Ugandans living with HIV-ImpLEmentation Strategies to Save lives (PULESA-Uganda), an NIH National Heart, Lung, and Blood Institute funded study evaluating the integration of HTN care within existing HIV services in HIV clinics in Uganda (NCT05609513). This mental health qualitative study was conducted during the formative phase of PULESA-Uganda, wherein 2 pilot sites were implementing integrated HIV/HTN packages. Clinic A is located within a public sector health facility in Wakiso district which also offers outpatient care, maternal and child health services, vaccination with separate HIV and HTN clinics. The facility is headed by a medical clinical officer (similar to advanced practice providers in the U.S.) and the staff include nurses, medical clinical officers, laboratory technicians, data officers, pharmacy technicians, and HIV peer counselors. Clinic B is located within a private not for profit (PNFP) hospital in Wakiso district. Although the hospital has specialist and non-specialist physicians, the HIV clinic team comprises medical clinical officers, nurses, laboratory technician and HIV peer counselors. HIV services at both clinics are supported by local PEPFAR implementing partners. HTN services were integrated into HIV care at both clinics during the PULESA-Uganda pilot phase. This meant that PLHIV were screened for hypertension, diagnosed, prescribed, and dispensed medication free of charge, and assessed for adherence and hypertension-related complications by the same healthcare providers and at the same points where they received HIV-related care.

Data collection procedures

We purposively sampled and recruited healthcare providers across both study sites to participate in a single elicitation qualitative interview exploring their perceptions on the feasibility of integrating mental health services within HIV care. Participants were eligible if they 1) had responsibilities in delivering services for HTN services for PLHIV and 2) had responsibilities in providing mental health services. Providers were excluded based on having professional responsibilities mostly related to storing and documenting patient data. Given the dynamic nature of HIV service delivery, where providers must step outside their designated roles to support understaffed areas of the clinic, we did not sample by job title.

The interview guide was developed by co-authors DG, AH, JIS, CB, and AK. The final guide included the following key domains: provider background and responsibilities, conceptualization and orientation to mental health, confidence in appraising mental health problems, and perceptions on organizational capacity for mental health service integration. Some of the interview questions were deductively derived from the Inner Setting and Outer Setting domains outlined in the Consolidated Framework for Implementation Research (CFIR) [17] and the Knowledge, Skills, Social/Professional Role and Identity, and Motivation and Goals domains from the Theoretical Domains Framework (TDF) [18]. Others were inductively constructed based on observations of clinics and insights from co-authors JBK, CK, JIS, FCS, and CL.

Data collection occurred between July 7 and August 3, 2022. The lead author (DG) conducted the interviews in English. All interviews were conducted at the providers’ place of work. DG audio taped the interviews and kept field notes at the end of each interview, tracking her positionality and summarizing key points from the participants’ narratives. Interviews were subsequently debriefed with the co-authors FA, JBK, CK, JIS, and FCS. These conversations helped explore the reliability and validity of the findings and helped contextualize findings within mental healthcare and HIV care in Uganda. On average, the interviews were 45 minutes in length, ranging from 30 to 60 minutes overall. The final interview guide can be found in S1 File.

Data analysis

Concurrent with and after data collection, DG transcribed and open coded the interviews. An initial iteration of the codebook was generated through research team discussions, with a combination of inductive codes emerging from the interviews and deductive codes rooted in the CFIR and TDF constructs. For example, inductively, we coded for cultural perceptions of mental health among the provider population and challenges providers encounter when interacting with patients’ family members. Deductively, we coded for inner and outer setting features of the HIV clinics as well as provider level knowledge and confidence around appraising mental distress. Co-authors DG and FA independently piloted the codebook with 6 interviews and discussed coding discrepancies and ideas that were not being captured. A discrepancy that emerged at this stage of the analysis was whether to code provider perceptions of mental health differently from providers’ understandings of mental health protocols within the clinic. The authorship team decided to code provider perceptions separately from provider understanding to identify nuances between individual understandings of mental healthcare and organizational capacity for delivering related services. Fig 1 below demonstrates which CFIR and TDF constructs were used in the analysis, and how they relate to one another to paint a picture of the larger implementation climate.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Relationship among CFIR and TDF Constructs Used in Analysis.

https://doi.org/10.1371/journal.pone.0353859.g001

After the coding agreement was achieved, DG independently coded the dataset using NVivo 20. The authorship team and field notes were regularly consulted to establish context. For example, CK provided insights on how patients assume somatic manifestations of anxiety and depression as similar to other physical health conditions. DG kept annotations and memos throughout the coding process to track analytical insights and emerging themes. Theme development centered around how providers adapt within resource constraints to address complex patient needs and fulfill clinical obligations.

Iterative coding and data debriefs concurrent with data collection led the research team to believe that data saturation was achieved. These procedures allowed the team to identify additional care delivery perspectives that were missing, like those of a laboratory technician, as well as implementation topics, like differentiated service delivery mechanisms.

While DG was the primary analyst formally coding the interviews, several strategies were employed to ensure analytic rigor and trustworthiness throughout the study. As mentioned, transcripts and initial open codes were collaboratively reviewed by members of the research team to iteratively refine the codebook. This ensured that DG applied each code consistently throughout the dataset, while maintaining analytic continuity as the primary coder. Data and coding debriefs with those who developed the codebook helped verify this as well. Given the interpretive nature of qualitative inquiry, the authorship team did not prioritize quantitative measures like Cohen’s kappa or interrater reliability. Contemporary qualitative methodologists have noted that such quantitative metrics do not always reflect interpretive approaches to thematic data, and that thematic analysis extends beyond multiple coders simply replicating a coding process. The authorship team instead emphasized the level of engagement with the data and transparency of analytic decision making throughout the coding process [19].

Statement of reflexivity

The study team consisted of Ugandan and American researchers, with varied backgrounds in public health, clinical medicine, and medical anthropology. Notably, the lead author and primary interviewer for this study is an Indian American who comes from a different cultural background from the study participants and context. This outsider perspective was balanced through extensive rapport building with the study team and participants, which helped challenge any underlying assumptions.

Statement of ethics

Written consent was obtained from all participants prior to their involvement in the study. All elements of the study, including the interview guide, methods, and participant consenting process, were ethically approved by the Institutional Review Boards at Yale University (New Haven, Connecticut, USA) (IRB Protocol ID: 2000032922) and Makerere University School of Medicine Research and Ethics Committee (Mak SOMREC) (Kampala, Uganda) (IRB Protocol ID: REF Mak-SOMREC-2021–58). The study also received clearance from Uganda National Council of Science and Technology (UNCST). All paper consent forms were kept in lockable cabinets only accessible to the study team and all audio-visual elements were stored in a data-encrypted, HIPAA compliant cloud storage system.

“Making it Work:” The invisible labor that overcomes institutional, staff, and patient challenges

It is important to note that participants frequently referenced hypertension integration efforts when discussing potential efforts to integrate mental health care, given their prior and current experiences with these efforts and the similar infrastructural demands required. Although hypertension is not a focus of the study, references to it will appear throughout the interviews as a lens through which the providers interpret mental health service integration. All the providers mentioned specific barriers and facilitators to delivering mental healthcare at the institutional level, the staff level, and at the PLHIV level. Mental health-specific clinical protocols were largely absent, and providers reported differential levels of exposure and training to mental health services. While some nurses received formal psychosocial training in their education, others recall never receiving formal or practical training. However, several resources were embedded in the HIV care system that providers drew upon to support patients with mental distress. These included peer counselors (also referred to as expert clients) who provided psychosocial support and follow-up services, as well as nurse counselors trained in adherence counseling, which often served as a mental health intervention. One clinic also had a part-time psychiatric clinical officer available on select days. This resource context greatly informed the providers’ ability to provide mental health care. Collectively, these barriers and facilitators paint a larger picture of the implementation climate for mental healthcare service integration within HIV clinics. They are highlighted in Table 2.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Institutional, staff, and patient level barriers and facilitators to MH service provision within HIV care.

https://doi.org/10.1371/journal.pone.0353859.t002

The providers identified these barriers and facilitators based on their experiences in adopting and integrating HTN care within HIV care. They perceived many of the barriers and facilitators to successfully integrating HTN as the same to successfully integrating mental health within HIV care. For example, supportive family structure is a patient-level facilitator that eases their ability to improve mental health and HTN health outcomes for PLHIV. Likewise, high workloads are a provider-level barrier to delivering integrated HTN/HIV and mental health/HIV care long term.

An important deviation, however, is the clinics’ logistical set up and documentation systems. The providers noted that the existing clinical infrastructure is sufficient to adopt HTN services, as they “can openly measure the patient’s blood pressure” (PID 9, Clinic B, Clinical Officer) without relying on private spaces, since HTN is less stigmatizing than mental health. For the purposes of mental health service integration, however, this clinical infrastructure is a barrier. They reported that mental health services require “more privacy to ask sensitive questions” (PID 7, Clinic B, Clinical Officer) which the clinics do not presently have. Our analysis reveals that the healthcare providers, in negotiating the HIV clinics’ context and patients’ context described in Table 2 to deliver mental healthcare services, are at the center of two negative feedback loops. These relationships are summarized in Fig 2.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. The Impact of “Making it Work” on HIV Healthcare Providers in Central Uganda.

https://doi.org/10.1371/journal.pone.0353859.g002

The first negative feedback loop is between the providers and the HIV clinics. The HIV clinics’ implementation features (time, protocol, and space limitations) shape the limited resources available to the provider to deliver mental healthcare services. The providers experience tension when they operate within these constraints to meet performance expectations and obligations to the HIV clinic. The second negative feedback loop is between the providers and their patient population. PLHIV present to the provider a wide range of psychosocial problems, from overt behaviors like excessive talking and alcohol dependence to more invisible mental health problems like depression and suicidal ideation. Providers negotiate these problems through adherence counseling, which is often limited to a select few PLHIV with poor adherence and is insufficient in detecting, addressing, and treating them. A clinic officer at Clinic B explains the cumulative impact of navigating these negative feedback loops:

“But it’s not your fault sometimes. You have to counsel yourselves that ‘it’s not my fault’. You have to share it with your team that you are trying your best, you are just trying to make it work.” (PID 7, Clinic B, Clinical Officer)

The remaining results will elucidate how the features highlighted in Table 2 interact with one another to create these negative feedback loops. It will also explore the process of how providers leverage clinic, staff, and patient level facilitators to “make it work,” highlighting the invisible labor they engage in to overcome the barriers highlighted in Table 2. TDF was used to make meaning of the provider and their own context, namely how their confidence and self-efficacy is impacted as they navigate these negative feedback loops. The CFIR was used to elucidate the inner setting (e.g., clinic infrastructure, protocols, and documentation systems), the outer setting (e.g., community stigma, resource availability, patient socioeconomic context), and implementation climate (e.g., readiness for change). This framework allowed us to situate provider narratives within broader contexts of the HIV clinics and patients’ lived realities.This deductive analysis was complemented with inductive analysis to understand providers’ experiences in “making it work:” the unrealistic demands placed upon both the providers and their patients and how these take a mental toll on the provider as well.

Providers strain professional boundaries to ensure PLHIV achieve viral suppression

Providers frequently contend with barriers on the HIV clinic level (insufficient private spaces, protocols, and comprehensive documentation platforms for mental health indicators), as well as barriers on the staff level like significant workloads. A negative feedback loop between the providers and the clinic emerges when these barriers complicate their ability to meet viral suppression goals for PLHIV with mental distress. The providers navigate this negative feedback loop by leveraging facilitators on the clinic level, like decentralized follow-up services.

Providers optimize ARV adherence counseling to respond to PLHIV’s mental distress

A key facilitator to the integration of mental healthcare service delivery in the provider context is that providers have vast interpretations of mental health problems among their patient populations. Because of this, they are perceptive to their patients’ comport and can identify when additional psychosocial support is needed. For example, some summarized their mentally distressed patients as those who exhibit behaviors like “disorganized speech, excessive talking, appearing confused, or dirty, or crying” (PID 10, Clinic B, Nurse Counsellor). Others mentioned encountering patients with more invisible mental distress, like “suicide, sad, or stress” (PID 2, Clinic A, Nurse Counsellor), which is not readily noticeable to the provider, unless indicated by the patient.

A key barrier in the clinic context, however, is the limited scope for psychosocial assessments within adherence counseling. A universal responsibility among the providers, adherence counseling is understanding “what the people are going through” (PID 1, Clinic A, Psychosocial Counsellor) that impedes their ability to adhere to their antiretroviral (ARV) medications. Typically, these interactions are understanding what behaviors are impacting the patient’s ability to adhere to ARVs and how they can be modified to achieve viral suppression. A negative feedback loop between the provider and the patients they treat emerges when the providers are perceptive to their patients’ mental distress, but do not have the adequate tools in the clinic to address it. Providers navigate this negative feedback loop by leaning on religion, empathy, and visualizing innovative solutions to “make it work” with adherence counseling.

Unrealistic organizational and cultural demands are barriers to mental health service provision

The providers noted that while navigating the negative feedback loops with the clinics and the patients, they recognize unrealistic expectations placed both on themselves. The nurse counselor at Clinic B mentioned an unrealistic demand placed upon her when she encounters patients experiencing mental distress:

“They [PLHIV] feel bad because they thought that we were going to provide food and transport. They thought we could cure HIV. They thought we could do these things… Even ‘fix bewitchment.’ [...] We cannot do those things. But they feel bad. We just counsel them, and they get caught up with the situation.” (PID 12, Clinic B, Nurse Counsellor)

The provider describes several sociocultural factors that shape a patient’s predisposition to mental distress that she is tasked with addressing. Oftentimes, patients request additional support for food and transport, have difficulties accepting the chronicity of their HIV status, and want providers to address their state of “bewitchment.” (Outer Setting, CFIR). These perceptions of bewitchment are common constructions of psychological distress in central Uganda and often lead to impacted individuals feeling ostracized. The nurse counselor notes how in these adherence counseling interactions, she can provide health education and explain the impacts of not adhering, and the patients reconcile with that. Along with having to address “bewitchment,” the psychiatric clinical officer at Clinic A discusses how he reckons with this label as well:

“Most of the people that come to see us are stigmatized. Sometimes, they remain in the community. They [patients with mental health problems] fear coming to us. Community calls them omulalu, and sometimes they call us, the psychiatric providers, omulalu. Even if you try to see them in their home, they will not open their doors for you.” (PID 4, Psychiatric Clinical Officer, Clinic A).”

Along with being bewitched, patients with mental health problems are construed and labeled as omulalu. It is a Luganda word encompassing observed behaviors that are considered or perceived to be socially abnormal. The term encompasses simple outbursts to more severe manifestations of mental illness. It holds a deeper significance as a stigmatizing word relating to madness (Outer Setting, CFIR). Fearing this socially isolating label, many patients suffering from distress do not openly seek care and resist coming to the clinic for support. This places the onus on the provider to actively seek these patients in the communities, which is difficult to achieve because the providers are not readily received for help and support. The psychiatric clinic officer also indicated that in providing mental health related support to patients, he himself has been termed as omulalu. He accepts such labels as “normal,” but acknowledges that he faces an unrealistic demand to resist these long-standing communal stigmas to provide mental healthcare.

Overall, providers indicated that operating within the fullest extent of their context and going above and beyond is sometimes not enough to facilitate positive mental health or holistic health outcomes. The unrealistic expectations on themselves and their patients imposes significant barriers to both provide mental healthcare and receive it. Sometimes, “you [the provider] can educate them, do all you can, and at the end they [the patient] will say ‘I will take my herbal medicines.’” (PID 3, Clinic A, Clinical Officer). If the patients feel that their needs are not met in the clinic, they seek alternative sources of care, and this can negatively impact their mental health problems and abilities adherence to ARV medications. Patients are often seeking non biomedical solutions to their circumstances, hindering a provider’s ability to provide adequate care..

The providers documented a variety of procedures that ease the complex sociocultural and organizational burdens placed upon them. The clinical officer at the public facility describes a stark change she noticed among her patients’ ability to open up to her after a medical doctor with HIV and HTN, who does not work at the health center, came to speak to the patients about his journey with these conditions. She described:

“The doctor told the patients, ‘If you have any problems, tell the clinical officer.’ When he expressed this to them, the patients were happy. When we received them, they opened up more, they said they will come back next time for more treatment. It’s important for them to have someone who understands their problems tell them that ‘it will be okay’.” (PID 3, Clinic A, Clinical Officer)

The clinical officer recognizes that a limitation she cannot overcome is shared experience with the patients who have a disease. After a reputed individual with HIV and HTN spoke to the patients about how they can live with their chronic conditions and how the healthcare providers can support them, this burden was lifted from the provider, and she could better tend to her patients’ emotional needs. The providers generally recognize that there are some patient needs they cannot meet through “making it work” without formalized integrated mental healthcare services. They are frequently tasked with addressing larger, systemic issues like poverty and cultural stigmas. The providers discuss how certain processes like relying on expert clients and facilitating social support mechanisms overcome these demands, but they feel uncomfortable to go the certain lengths they do to leverage these mechanisms.

Providers lose it as a consequence of “making it work.”

HIV healthcare providers, in “making it work” and while experiencing unrealistic demands in doing so, describe themselves as “basically depressed and traumatized” (PID 6, Clinic B, Nurse Counsellor). They discuss how oftentimes; they know and understand the appropriate course of action to take when they encounter patients experiencing mental distress but feel constrained in their ability to do so. For instance, a nurse counselor at Clinic B indicates:

“Someone can say to you: ‘leave me to die.’ That person will not listen to anything. ‘Leave me to die.’ If you hear those statements, you know that you must put more effort there. That is just one sentence, but it means a lot. In your heart, you don’t feel good. You know there is not much you can do for them.” (PID 10, Clinic B, Nurse Counsellor)

The provider describes feeling a sense of dread whenever she hears patients wanting to be left to die (Anticipated Regret, TDF). The nurse counselor understands the impact an HIV diagnosis has on the patient’s own psyche and reflects on the fact that she is limited in her capability to address the mental distress. Patients showing signs of suicidal ideation require many additional efforts to probe into their mental health, and as previously mentioned, these additional efforts are beyond their cadre, training, ability. Providers additionally documented these experiences long after their interactions with their patients as well.

“There was this lady, she came to test for HIV. She was raped by her boss. And then she was told to not tell anyone. Then she was like Omusawo [healthcare provider] ‘why is it that that guy raped me?’ That lady was like losing it. We couldn’t manage her because she moved away. We can’t afford to look for her. I lose it when I think about her.” (PID 14, Clinic B, Laboratory Technician)

The provider describes living with the guilt that she could not do more for her patients considering the constraints that she is working within. She describes internalizing the psychosocial pain that the patient presents. Within the context of limited resources, the provider knows that there is not much she can do for the patient and lives with that guilt every day.

As providers cycle through these negative feedback loops, they experience the cumulative deprivation of resource limitations, feel constrained in their abilities to comprehensively tend to patient needs, and internalize their patients’ pain as well. This results in the depletion of their own context. So much so, that they want to “leave healthcare, and go and do other things.” (PID 6, Clinic B, Nurse Counselor). Providers frequently “go home exhausted, come back [to the clinic] exhausted, and leave again exhausted” (PID 1, Clinic A, Psychosocial Counselor). In the absence of formalized mental health services, the providers find themselves in perpetual, cyclical, context-depleting cycles that impact their own quality of life.

Facilitating a positive clinic and provider feedback loop

Our findings revealed that HIV providers experience a negative feedback loop with the HIV clinic because of the few infrastructural resources at the clinic level available to cater to PLHIV’s mental health needs. These include limited mental service delivery protocols, platforms where mental health indicators are tracked and monitored along the care continuum, and especially in public facilities, limited private spaces to ask sensitive mental health related questions. These barriers culminate and make it difficult for providers to ensure their PLHIV with mental distress achieve viral suppression, a key performance metric for the provider and the clinics at large. Providers adapt within this context by performing patient outreach services outside of the health facility on their own time, money, and resources to achieve their professional responsibilities. Facilitating a positive clinic and provider feedback loop, one which provides adequate clinic level resources to address mental health needs, is vital to successfully integrating mental healthcare. In terms of enhancing clinic specific guidelines or protocols on mental health services integration, prior research in Uganda suggests that a protocolized clinical depression screening model, in which validated tools are used for assessing mental health, yielded more accurate assessments of depression than relying on subjective clinical acumen [20]. Integrating these assessments within HIV/HTN and other care cascades can provide technical support in addressing patient mental health needs and holistically support PLHIV outcomes, without straining provider resources.

Facilitating a positive patient and provider feedback loop

Providers are perceptive to PLHIV’s mental distress; however, they feel ill-equipped to address this distress in the clinical setting. The most salient feature in this setting to address mental distress is adherence counseling, a practice by which nurse counselors work one-on-one with patients to understand what challenges patients face in adhering to ARVs. Provider narratives indicate that adherence counseling is sometimes insufficient in appraising deep emotional distress like suicidality. Sometimes, providers who have the capacity to, overcome the limitations of adherence counseling with various inventive problem-solving strategies, that if protocolized, can aid them in their diagnostic efforts. Previously in Uganda, a savings-led family based economic empowerment intervention has successfully improved adolescents linkage and adherence to ART [21]. Our findings corroborate that financial insecurity is a determinant of mental distress, and providers address it by augmenting adherence counseling with economic empowerment recommendations like home gardening. Protocolizing these practices that providers already engage in can support mental health integration efforts. Additionally, prior literature suggests that culturally relevant psychosocial programming for PLHIV, especially ones with racially congruent images and framing, are more effective in improving mental health outcomes [22]. Including religiosity in such culturally relevant programming can especially aid both PLHIV and their providers in improving mental health outcomes.

Additionally, interdisciplinary efforts have shown promising success in improving PLHIV mental health needs. A screening integration effort in Zimbabwe capitalizing on nurses, community health workers, traditional medicine practitioners, and routine HIV service providers in clinics reported a more balanced work load among the care team. They also reported more streamlined uptake of referral pathways and improved relationships with their patients [23]. This further highlights our findings that when there is cohesion among clinical systems and the providers, relationships with patients improve and mental health needs can be more adequately cared for.

Overcoming the insurmountable conditions of the clinic-provider-patient negative feedback loops

While navigating these negative feedback loops with the clinic and the patients, the providers note unrealistic demands placed upon both them and their patients. First, they describe having to contend with longstanding communal and interpersonal mental health stigmas. They note that culturally constructed mental health stigmas prevent patients from seeking mental health specific care, and that many stigmatize those who provide formalized mental health services as well. Stigma reduction approaches must be considered in integrated HIV and mental health care. The providers have relied on peer clients to appeal to patients to talk about their chronic conditions openly and honestly with them, and this has showcased some success. In Botswana, a stigma reduction approach through leveraging peer counselors to establish rapport reduced the burden on both the healthcare provider and the HIV positive patient [24]. Additionally, other peer interventions have shown high acceptability among young mothers with HIV in Uganda, especially with mental health related issues. Our findings demonstrate that peers or expert clients are critical in linking PLHIV to integrated HTN care, and these facilitators can be leveraged for mental healthcare as well.

Adapting within these negative feedback loops poses a significant toll on the providers’ own mental health. They describe feelings of longstanding, residual guilt after being unable to provide comprehensive care to their patients. A lack of formalized mental health services has resulted in patients losing contact with their providers and in some cases, passing away. Prior ethnographic research in Uganda suggests that nurses and other healthcare providers, in the absence of adequate resources to provide comprehensive care to patients, experience moral distress. Despite knowing and wanting to take ethically appropriate actions, the providers are limited in their ability to actually do so [25]. This is particularly true in the HIV context, as nurses and other providers pull from their personal resources to cater to complex patient needs, and vicariously experience the long-term impacts of not meeting them.

Implementation strategies for integrated mental health and HIV services can leverage the providers’ innate desires to want to help their patients overcome these needs and institute the appropriate supervision systems and mental health support to achieve these goals. Specifically, strategies such as provider training in mental health screening and communication, ongoing supervision and mentorship from specialists, routine feedback loops, and integrating screening tools into HIV care workflows can help institutionalize what providers are already engaging in formally. These strategies build on providers’ existing commitment and resourcefulness, and help scale sustainable, low resource approaches to delivering mental health support without requiring entirely new interventions. When incorporated within clinical settings and existing integrated models of care, these strategies can create sustainable environments for providers to deliver comprehensive mental healthcare.